ARTICLE IN PRESS
Planetary and Space Science 55 (2007) 2164–2172
www.elsevier.com/locate/pss
Local plasma processes and enhanced electron densities in the lower
ionosphere in magnetic cusp regions on Mars
E. Nielsena,, M. Fraenza, H Zoua, J.-S. Wanga, D.A. Gurnettb, D.L. Kirchnerb,
D.D. Morganb, R. Huffb, A. Safaeinilic, J.J. Plautc, G Picardid, J.D. Winninghame,
R.A. Frahme, R. Lundinf
a
Max Planck Institute for Solar System Research, 37191 Katlenburg-Lindau, Germany
Department of Physics and Astronomy, University of Iowa, Iowa City, IA 52242, USA
c
Jet Propulsion Laboratory, Pasadena, CA 91109, USA
d
Infocom Department, ‘‘La Sapienza’’, University of Rome, 00184 Rome, Italy
e
Southeast Research Institute, P.O. Drawer 28510, San Antonio, TX 78228-0510, USA
f
Swedish Institute of Space Physics, Box 812, S-98128 Kiruna, Sweden
b
Received 24 January 2007; received in revised form 3 July 2007; accepted 6 July 2007
Available online 26 July 2007
Abstract
Both the MARSIS ionospheric sounder and the charged particle instrument package ASPERA-3 are experiments on board the Mars
Express spacecraft. Joint observations have shown that events of intense ionospheric electron density enhancements occur in the lower
ionosphere of magnetic cusp regions, and that these enhancements are not associated with precipitation of charged particles above a few
hundred electron volts (o300 eV). To account for the enhancement by particle precipitation, electron fluxes are required with mean
energy between 1 and 10 keV. No ionizing radiation, neither energetic particles nor X-rays, could be identified, which could produce the
observed density enhancement only in the spatially limited cusp regions. Actually, no increase in ionizing radiation, localized or not, was
observed during these events. It is argued that the process causing the increase in density is controlled mainly by convection of ionosphere
plasma driven by the interaction between the solar wind and crustal magnetic field lines leading to excitation of two-stream plasma waves
in the cusp ionosphere. The result is to heat the plasma, reduce the electron–ion recombination coefficient and thereby increase the
equilibrium electron density.
r 2007 Elsevier Ltd. All rights reserved.
Keywords: Mars; Ionosphere; Convection; Sounding radar
1. Introduction
The topside radio wave sounder MARSIS (Mars
Advanced Radar for Subsurface and Ionosphere Sounding) (Picardi et al., 2004) on board the Mars Express
spacecraft (Chicarro et al., 2004) measures reflections from
electron densities in the Martian ionosphere as a function
of frequency and radio wave propagation time (Gurnett
et al., 2005). Several different kinds of targets giving rise to
reflections have been identified, and a variety of them are
poorly understood. Single and double peaks in the vertical
Corresponding author. Tel.: +49 5556 979450.
E-mail address: [email protected] (E. Nielsen).
0032-0633/$ - see front matter r 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.pss.2007.07.003
profile of electron density have been observed, oblique
echoes from upward bulging ionosphere in cusp regions
(see Duru et al. (2006); Nielsen et al. (2007)), diffuse echoes
from off-nadir directions, and possible holes in the
ionosphere are indicated. The most common observation
is vertical reflections from nadir. Electron densities in the
Martian ionosphere are generally horizontally stratified
and the dominating observation by the sounder is vertical
reflections from these layers. Techniques have been
developed whereby the observations are inverted to the
vertical profile of electron densities (Nielsen et al., 2006).
The primary density peak in the profile conforms to the
predictions of photochemical equilibrium between ionization and recombination in a plasma. For example, the peak
ARTICLE IN PRESS
E. Nielsen et al. / Planetary and Space Science 55 (2007) 2164–2172
density conforms to a predicted spatial variation controlled
by the solar zenith angle (Chapman, 1931; Gurnett et al.,
2005; Nielsen et al., 2006). However, at times sudden short
lasting strong increases are observed in the maximum
electron density in the primary density peak. This separates
the events from the overall behavior of the background
ionosphere observed along the spacecraft trajectory. Such
intense events have been mentioned before (Gurnett et al.,
2007) and are analyzed in this work invoking charged
particle measurements by the particle detector ASPERA-3
(Analyzer of Space Plasmas and Energetic Atoms-3) on
board Mars Express (Barabash et al., 2006; Lundin et al.,
2004). An absence of extra ionizing radiation during these
events points to local processes in the ionosphere as the
cause for the increase in electron density.
2. Observations
During ionosphere soundings with MARSIS echoes are
observed for a series of discrete frequencies (160). The
Case1: Orbit 2304
Nem(105cm−3)
5
4
4
3
3
3
2
2
2
1
1
1
Normal
Event
20:45
20:52
10−2
X−ray flux(Wm−2)
Case3: Orbit 2491
5
20:38
21:00
1−8A
0.5−4A
10−3
10−4
10−5
10−6
10−7
06:00 06:07 06:14 06:21 06:28
X
M
C
B
A
10−2
10−3
10−3
10−4
10−6
10−7
10−9
10−9
20:45
20:52
21:00
X
M
C
B
A
10−5
10−8
20:38
04:33
10−2
10−8
10−7
10−8
06:00 06:07 06:14 06:21 06:28
10−9
4
3
3
3
2
2
2
1
1
1
20
40
60
SZA(°)
80
0
20
40
60
SZA(°)
80
04:33
04:48
05:02
UT:2005/12/21 04:29:47−−05:09:38
4
0
X
M
C
B
A
10−6
5
Normal
Event
Best−fit
05:02
10−5
5
4
04:48
10−4
UT:2005/11/14 05:57:22−−06:33:04
UT:2005/10/29 20:32:16−−21:02:02
5
Nem(105cm−3)
sweep through frequencies last 1.2 s and the measurements are repeated with 7.5 s intervals. The maximum
signal frequency for which an echo is recorded (before the
radar signal passes through the ionosphere without
reflection) is here taken as a measure of the peak plasma
frequency (electron density) in the ionosphere primary
layer. In Fig. 1, observations during three different orbits
are displayed in three vertical stacks of panels. The top
panel in each stack is the peak density versus Universal
Time. During each orbit there is a component of the
maximum density slowly varying with time along the
trajectory. This slow variation is occasionally interrupted
by short lasting strong increases of the density. Electron
density enhancement occurred during one time interval of
orbit #2304, on October 29, 2005. In both orbits #2359
from November 14, 2005, and #2491 from December 21,
2005, there are three intervals with strong deviations. The
mean increase in the excursions is on average about 50%
over background. The altitude of the peak density in the
excursions is o125 km. The time duration of the strong
Case2: Orbit 2359
5
4
2165
0
20
40
60
SZA(°)
80
Fig. 1. Strong peak electron density enhancement events observed during three orbits. The maximum density shown as function of Universal Time (top
panels) and solar zenith angle (lower panels). X-ray flux at GOES versus time showing absence of solar activity during the observations (center panels).
ARTICLE IN PRESS
2166
E. Nielsen et al. / Planetary and Space Science 55 (2007) 2164–2172
increases was less than 1 min. During these measurements
the spacecraft was at low altitudes near 300 km where the
spacecraft is moving rapidly with a speed of 5 km/s.
An event lasting a short time when observed from a
moving spacecraft can be either an impulsive event in time
or indicate a small-scale spatial structure rapidly crossed by
the spacecraft. It has been shown before that impulsive
solar X-ray bursts are causing sudden increases in the
Martian electron densities (Mendillo et al., 2006; Nielsen
et al., 2006). In the middle panels of Fig. 1 the solar X-ray
flux from 1 to 8 Å observed on the GOES (Geostationary
Operational Environmental Satellite) spacecraft are shown
for the times of MARSIS events (Bornmann et al., 1996).
Clearly, no solar X-ray activity was detected during these
times. Solar cosmic rays are also known to influence the
ionosphere densities (Morgan et al., 2006). Examination of
charged particle fluxes observations on board ACE
(Advanced Composition Spacecraft) and on board SOHO
(SOlar and Heliospheric Observatory) during the times of
the events, revealed no enhanced energetic charged particle
activity. GOES, SOHO and ACE are located near the
Earth, which was separated from Mars by an azimuth
angle of 29.41 (#2491), 6.21 (#2359), and 7.91 (#2304) (with
the orbit number in parenthesis). Neither at the time of the
events, nor when taking into account co-rotation time and
radial distance differences from the Earth to Mars, could
large-scale sporadic energetic precipitation be found to
account for the measurements. It follows that the events
are more likely associated with spatially limited processes.
In the lowest panels in Fig. 1 the peak densities are
shown as function of solar zenith angle. The background
densities are smoothly varying and well-fitted by a curve
reflecting the theoretical variation of the peak density
versus zenith angle in a Chapman layer. The events are
observed between 201 and 601 zenith angle, i.e. well into the
dayside ionosphere. The duration of the events and the
speed of the spacecraft combines to indicate that the spatial
scale of the enhancements was o300 km. This corresponds
to about 51 in latitude. The only structures on Mars, which
may have effects in the ionosphere with such a scale, are the
crustal magnetic fields. Crustal magnetic fields on Mars
were observed by MGS (Mars Global Surveyor); they
maximize in the southern hemisphere in the region
bounded by 130–2501 in east longitude and 851 to 101
latitude (Acuna et al., 1998, 1999; Connerney et al., 1999).
Magnetic field measurements in 400730 km (Connerney
et al., 2001) were used to determine the global variations of
magnetic field inclination and magnitude. The result in the
region of strong fields is shown in Fig. 2. The top panel in
Fig. 2 shows the magnetic inclination. ‘Blue’ color is
upward-pointing vertical field, and ‘red’ demarks downward-pointing fields. There is a strong spatial variation,
especially in latitude, with oppositely directed fields
alternating with each other, separated by regions of closed
magnetic field lines (‘green’). In the lower panel, the
magnitude of the magnetic field shows a single strong
maximum. Superposed on this are the ground tracks for
the three orbits under investigation, with the locations
marked at which events with strong density enhancement
were observed. It reveals a clear pattern in which the events
occur in magnetic cusp regions with vertical magnetic field
lines (with upward- or downward-pointing magnetic
fields). Further, the events were observed in the region of
strongest crustal magnetic fields.
With the location of enhanced densities limited to the
cusp regions, local acceleration of charged particles somewhere along the magnetic field lines may be a possible
cause of the density increase (Lundin et al., 2004).
Fortunately, the ASPERA-3 particle experiment was in
operation during these events. The ASPERA-3 instrument
is described in Barabash et al. (2006). The instrument
consists of an electron spectrometer (ELS), an ion spectrometer (IMA) and three neutral particle sensors. Fig. 3 (top
panel) shows a spectrogram from the ELS observed during
orbit #2359. The energy range of the sensor is 0.4–20 keV,
but at this time, electrons with energies below 5 eV are
reflected to avoid saturation of the counters. Shown are
counts obtained during 4 s sampling intervals by all 16
anodes of the sensor. After 0629 UT, the spacecraft enters
the magnetosheath, and before that the spacecraft is inside
the photoelectron boundary. We observe that there are no
significant fluxes of electrons above 300 eV. Specifically, no
energetic electron flux coincided with the ionosphere
electron density increases observed by MARSIS (Fig. 1).
The ELS sensor can roughly estimate electron bulk
velocities using 16 anodes mounted in a plane. This
estimate indicates radially outward flows for all flux
enhancements. The IMA sensor observes no ions with
energies above 300 eV; specifically no protons. For periods
of less than 12 s, heavy ion flows with energies of less than
200 eV/charge are observed. The flow speed of these ions is
lower than 10 km/s and directed outward. The middle
panel shows similar electron data for orbit #2491. Also, for
this orbit there are no electrons above 300 eV and only lowenergy ions associated with the electron flux increases at
10–300 eV. The bulk velocities were directed outward. The
crucial observation in our context is that there were no
energetic electron or ion fluxes observed during the entire
time interval including the time when electron density
enhancements were observed in the lower ionosphere.
Furthermore, the bulk velocity of the observed electrons
and ions were directed outward, i.e. there was no
precipitation of higher energy charged particles into the
ionosphere. The spatial locations, where these excursions
of electron fluxes into higher energies up to 300 eV
occurred, are shown in the lower panel on top of a map
of crustal magnetic field inclination. The spatial locations,
where the more energetic electrons were observed, are only
marked where the magnetic field magnitude is large
(4100 nT; so as to minimize solar wind effects) and the
spacecraft altitude is below 500 km (near the altitude for
the Connerney-model magnetic fields). The low-energy
electron fluxes tend to be located in the cusp regions.
However, the correlation is not as clear as for the density
ARTICLE IN PRESS
E. Nielsen et al. / Planetary and Space Science 55 (2007) 2164–2172
2167
Fig. 2. Spatial maps of the inclination (top panel) and magnitude (lower panel) of the Martian crustal magnetic fields (Connerney et al., 2001). Superposed
on both maps are the ground tracks of the relevant orbits with locations marked where peak density enhancements were observed. There is a clear relation
of the enhanced density events with vertical magnetic field and with the magnitude of the field.
enhancements. This indicates that low-energy electron
fluxes tend to be more smeared out across the magnetic
field. The exact spatial coincidence of 300 eV electrons and
cusp regions remains tentative. The main reason for this
‘smearing’ may be that the spatial variations of the crustal
magnetic field is modified during these events (occurring on
the dayside) by a magnetic field component induced from
the interaction of the solar wind with the ionosphere. This
would directly influence the spatial distribution of the low
energy electrons.
Detrick et al. (1997) determined the minimum altitude an
electron can penetrate into the Martian atmosphere before
being stopped. With peak density enhancement located
below 125 km, precipitating electrons with energies 41 keV
are required. Thus, no energetic charged particle precipitation was observed, which could possibly account for the
increase of the density maximum. The near-vertical
magnetic field lines that connect to the nadir region where
the sounder measurements were made, did not carry any
energetic particles which could conceivably increase the
ionization rate in the lower ionosphere to the point of
accounting for the observed density enhancement.
Fig. 4 shows two MARSIS spectrograms taken on
November 14, 2005 (orbit #2359): panel ‘a’ shows the
observed echoes outside the cusp regions and panel ‘b’,
observations in the cusp. The outstanding difference
between observations in the two magnetically different
regions is the increase of the maximum frequency in the
cusp: from 3.2 to 5 MHz; an increase of a factor 1.5. This
is one of the largest increases observed and corresponds to
an increase in the maximum electron density by a factor
2.2. The mean value of the density increase (Fig. 1) is
close to a factor 1.5. The observations are inverted to yield
the vertical electron density profiles following the procedure outlined in Nielsen et al. (2006). The result is shown in
Fig. 5. The solid curves (marked 141 and 132, respectively)
are the inverted profiles corresponding to the data in the
left-hand and right-hand panels in Fig. 4. The two
dashed curves are Chapman layer fits to the density
profiles around the density maximum, and the dotted
ARTICLE IN PRESS
2168
E. Nielsen et al. / Planetary and Space Science 55 (2007) 2164–2172
Fig. 3. ASPERA-3 electron flux observations during two of the relevant orbits are in the two upper panels. Note the occasional flux enhancements of
electron energies o300 eV. The locations of flux enhancements are plotted on the map of magnetic inclination shown in the lower panel. The
enhancements tend to occur in cusp regions of vertical crustal magnetic fields coinciding with enhanced densities.
curve is the difference between the cusp and outside-cups
profiles. The largest density increase for this event was near
the peak altitude of 110 km.
3. Discussion
The density profile around the density peak is determined by photochemical equilibrium between ionizing
radiation and recombination of electrons and ions. In an
equilibrium situation, the electron density is proportional
to the square root of the ratio of electron–ion production
(by ionizing radiation) and ion–electron recombination
coefficient.
One way to increase the equilibrium density would be to
enhance the intensity of the ionizing radiation keeping
everything else constant. However, no enhancement of solar
activity in the form of solar energetic radiation or
precipitation of solar cosmic rays occurred, and no local
accelerations of charged particles were observed which
could produce the enhanced densities. The particle detector
observed only an increase in the very low energy electron
flux directed away from the planet. Thus, neither large-scale
‘global’ nor small-scale ‘local’ enhanced precipitation of
ARTICLE IN PRESS
E. Nielsen et al. / Planetary and Space Science 55 (2007) 2164–2172
2169
Fig. 4. Spectrograms of MARSIS observations outside (panel a) and inside (panel b) a cusp region. Note the increase of the largest reflected frequency
(maximum peak density) on entering the cusp.
260
240
220
Altitude(km)
200
180
160
140
Peak electron density: 178886.6cm−3
Peak altitude: 112km
120
141
132
100
80
60
101
102
103
Electron
104
105
106
density(cm−3)
Fig. 5. Vertical electron density profiles corresponding to the observations
shown in Fig. 4. The dashed curves are fits of a Chapman function to the
low-altitude measurements. The dotted curve is the difference between the
two density profiles and represents the enhancement of electron densities
in the cusp region.
energy into the Martian ionosphere can account for the
events. The ionization production term appears to be
constant.
Another possibility for enhanced electron density is a
reduction of the recombination coefficient a. A doubling of
the equilibrium density would require a fourfold decrease
of the recombination coefficient. The dominant ion at the
altitudes near the density maximum is O+
2 . For electron
temperatures below 1200 K, the dissociative recombination
rate for that ion is a ¼ 1U9 107(300XTe)0U7 (Schunk and
Nagy, 2000)U Thus, an increase of the density by a factor
between 1.5 and 2 implies an increase of the electron
temperature by a factor between 3.2 and 7.2. If the electron
temperature outside the cusp was 200 K, then in the cusp
ionosphere it would have to increase to between 600 and
1500 K. This considerable heating does not come from
energetic precipitation from the Sun or from acceleration
processes along the magnetic field lines, but must be
produced by local plasma processes.
In the Earth’s ionosphere, considerable heating of the
electron gas by plasma wave activity is a well-known
phenomenon (Dimant and Milikh, 2003; St.-Maurice,
1990; St.-Maurice et al., 1990). Charged particle motion
in the ionosphere is basically governed by an imposed
electric field (E), by gyro motion in a magnetic field (B) and
by collisions with other particles. Above 130 km in the
Earth’s ionosphere, both electrons and ions are E B
drifting, suffering few collisions. Below 80 km, the
motion of both species are controlled by collisions. In the
altitude range from 80 to 130 km, electrons are E B
drifting while ion motion is controlled by collisions. When
the relative bulk speed of electrons and ions exceeds the ion
acoustic velocity, plasma waves appear as the two-stream
instability is excited (or Farley–Buneman plasma instability; Fejer and Kelly, 1981). In the auroral zone, where
strong convection electric fields often occur, the relative
electron–ion speed and electron density peak maximize in
the E-region, where the ion acoustic velocity is about
400 m/s. Typically, plasma waves are observed when the
ionospheric electric field exceeds 20 mV/m (corresponding
to an E B speed of 400 m/s) when electrons are driven
with supersonic speed through the collision-dominated
ARTICLE IN PRESS
2170
E. Nielsen et al. / Planetary and Space Science 55 (2007) 2164–2172
ions. Radar observations have demonstrated strong
enhancements of electron temperature in the E-region
during strong electric field events. It was concluded that
the temperature of the electron gas is enhanced when the
convection electric field is strong enough to drive the
electrons through the ion gas at supersonic speed.
St.-Maurice et al. (1981) reported increases of electron
temperatures from 300 to 1500 K. The conventional
frictional (Joule) heating caused by the DC electric field
cannot account for the observed high electron temperatures. It is now generally accepted that the strong electron
temperature increase is anomalous and caused by the
turbulent electric fields in Farley–Buneman plasma waves.
That enhanced electron temperatures are associated with
excitation of the two-stream instability is well documented
(Foster and Erickson, 2000; Nielsen and Schlegel, 1985).
Nielsen and Schlegel (1985) examined the relationship
between coherent phase velocity measurements in the
E-region with electron drift velocities in the F-region
(electric field) and found the phase velocity to be limited
to a value near the ion acoustic velocity. They also found
this limiting phase velocity to be a strongly increasing
function of the electric field. This increase of the ion
acoustic velocity with electric field magnitude was interpreted as heating of the E-region electron gas by the
unstable two-stream plasma waves. Robinson (1986) has
shown that there is an ample source of energy in unstable
waves resulting from excitation of the two-stream instability, and found that the anomalous electron temperatures in
the disturbed high latitude E-region can be quantitatively
explained in terms of heating by plasma waves. The
predicted variation of ion acoustic velocity and electric field
showed a weak dependence on DC Joule heating but a
strong dependence on heating of the gas by unstable waves.
Sudan et al. (1973) proposed to describe the anomalous
heating of the electron gas by replacing the electron-neutral
collision frequency with an anomalous collision frequency.
Nielsen (1986) found the anomalous frequency to be a
factor between 4 and 6 larger than the normal frequency.
Since plasma heating is proportional to the collision
frequency, this increase is associated with strongly enhanced heating.
Heating of the electron gas leads to a decrease of the
recombination coefficient and an increase of the electron
density. While the heating is well established, the density
increase is less well established. But recently, Milikh et al.
(2006) reported electron density and temperature observations in the Earth’s high latitude ionosphere during heating
events. They found anomalous electron heating to be
localized between 100 and 120 km, where the magnetized
electron bulk speed exceeds the acoustic speed of the
unmagnetized ions leading to the Farley–Buneman instability. Because the electric fields were strong (4100 mV/
m, corresponding to 42000 m/s E B drift speed) the
instability resulted in strong anomalous heating. Heating
from a background electron temperature of 500 K up to
2500 K was observed. The associated enhancements in the
electron density were observed to be up to a factor 2, or
near a doubling of the electron density between 100 and
120 km. The electron density perturbations showed a good
agreement with the model that accounts for suppression of
the electron/ion recombination rate due to anomalous
electron heating.
The conditions for particle motion and excitation of the
two-stream instability near the primary electron density
peak in the Martian ionosphere are quite similar to the
conditions in the Earth’s E-region. For a magnetic field
amplitude of 1000 nT, the electron gyro frequency is
28.4 kHz, which equals the electron collision frequency in
90 km altitude. Thus, electrons are magnetized above
90 km height. Similarly, the ions are unmagnetized below
an altitude of 150 km where the gyro frequency of the
dominating ion and its collision frequency have the same
magnitude (0.5 Hz). In this altitude range, there is a
relative motion between electrons and ions. For a
sufficiently large electric field the speed of the electrons
will exceed the ion acoustic speed and excite the two-stream
instability resulting in heating of the electron gas.
The driving force in this scenario is the electric field. It is
suggested that the electric field arises owing to the flow of
the solar wind plasma past the crustal magnetic field cusp
structures. The magnetic field lines in the cusps extend into
the solar wind, and the flow of a conducting medium over
the field lines induces an electric field, such that the
magnetic field is frozen into the plasma and carried along
by plasma motion. The density enhancement events were
observed in cusp regions located near the largest crustal
magnetic fields on Mars. These strong magnetic fields can
reach further into the solar wind than field lines from other
regions, an indication that the solar wind interaction with
the crustal fields plays a role in forming the events. For a
solar wind speed of 200 km/s and a magnetic field
magnitude of 1 nT, the induced electric field in the solar
wind is 200 1000 109 1000 mV/m ¼ 0.2 mV/m.
Assuming the magnetic field lines are electric equipotentials, the electric field in the ionosphere is increased owing
to convergence of the field lines with decreasing altitude:
assume an increase of a factor 20–4 mV/m. Let the
ionosphere magnetic field be 500 nT and the E B drift
speed of the electrons is then 4 103/(5 107) m/
s ¼ 8000 m/s. The ion acoustic velocity in an O+
2 ion gas
with TiTe200 K is 450 m/s. The electron speed exceeds
the ion acoustic velocity by a large factor. Thus, excitation
of the two-stream plasma instability in the Martian
ionosphere is a realistic possibility.
Heating of the electron gas is controlled by the electric field
magnitude and by the Pedersen conductivity, which is mainly
determined by the electron density and the collision frequency
(Withers et al., 2005). The electron density typically peaks at
altitudes between 90 and 150 km in a region where the
electrons are magnetized and the ions unmagnetized. There is
a DC component from the applied electric field and an AC
component from the turbulent wave electric field. The DC
energy deposit in the ionosphere is partly controlled by the
ARTICLE IN PRESS
E. Nielsen et al. / Planetary and Space Science 55 (2007) 2164–2172
DC electric field and the height integrated Pedersen
conductivity, S2 N e H/B. For a density of
N2 105 el/cm3, a scale height H12 km and a magnetic
field as above, we find the conductivity to be 1540 S. For a
4 mV/m electric field, the heating is 25 mW/m2. This is
compatible with typical values for the Earth high latitude
ionosphere; with S10 S and E50 mV/m, the heating is
25 mW/m2. Even though the electric field we have inferred
for Mars is small compared to the fields in the Earth’s high
latitude ionosphere, the DC Joule heating at Mars is
compatible to the heating in the Earth ionosphere. The
reason for this is the large (height integrated) Pedersen
conductivity, or equivalently the large electron-neutral
collision frequency in the Martian CO2 atmosphere; the
frequency is 100 times larger than in the Earth’s nitrogen
atmosphere. However, at the Earth (and presumably at
Mars) this conventional frictional heating cannot account
for the electron heating. The increase in electron temperature
is widely accepted to be caused by the turbulent (AC)
electric field generated by the two-stream plasma instability.
The influence of the turbulent electric fields is to increase
the effective electron collision frequency by a factor of 4–6.
Since the Pedersen conductivity is proportional to the
collision frequency, we also expect the wave heating to
increase by such a factor 5 to further raise the electron
temperature.
It appears that the cusp ionosphere with a solar wind
imposed electric field of a few mV/m is unstable to the twostream instability. Owing to the large collision frequency of
electrons with a neutral CO2 gas, even a modest electric
field imposed by the solar wind can cause heating
comparable to heating at the Earth, where the Pedersen
conductivity is smaller but the electric field is stronger. The
Pedersen conductivity is proportional to the electron
collision frequency which is increased to an anomalous
level by the turbulent wave electric fields. This may result in
energy deposits of the order of 100 mW/m2 and result in
enhancement of the electron gas temperature.
If the temperature enhancement is taken to be proportional
to the energy deposition, then the density is expected to
increase by a factor between 1.6 and 1.9 (for an energy deposit
of a factor 5 larger than the DC deposit, the recombination
coefficient decreases by a factor 50.7 and the density increases
by (50.7)1/2 ¼ 1.7). It is realistic to expect density enhancements
by a factor up to 1.5–2.0. If the electron gas temperature
changes suddenly from T0 to T1, it will take some time for the
densities to change to the corresponding equilibrium
values. Milikh et al. (2006) determined the time constant to
be t ¼ 1/(2 (a1 ao)1/2No). For a five-fold temperature
increase and an initial electron density of 105 el/cm3 we find
t ¼ 44 s. Thus, within a few minutes the system has reached
more than 90% of its equilibrium value.
To suggest that heating of a plasma in a flux tube leads
to a density increase may seem counter-intuitive. To
preserve pressure balance, one would expect the heating
to be associated with an (upward) expansion of the heated
plasma. This is actually what ASPERA-3 observed. We
2171
suggest the upwelling is inhibited and limited by ionneutral collisions. This may reduce the upwelling so
effectively as to lead to an increased density. A condition
for the density increase is that the recombination coefficient is strong enough that it overwhelms the upwelling
process. A simulation of the heated plasma in a flux tube is
planed to check this model.
The predicted quite high electron drift velocities in the
ionosphere may give rise to ionization effects if they exceed
the Critical Ionization Velocity (CIV), i.e. if the electron
kinetic energy, 0.5 me v2CIV, exceeds the CO2 ionization
energy of 35 eV. However, in this case the CIV is much
larger (43.6 106 m/s) than realistic drift speeds. The
magnetized ions do not take part in this process.
In magnetic cusp regions where crustal magnetic field
lines extend far into the solar wind, it appears that an
electric field is induced so that the magnetic field lines are
carried along with (‘frozen into’) the plasma. Mapping of
the induced convection electric field along equipotential
magnetic field lines results in an electric field of a few
mV/m impressed on the ionosphere. An electric field of
such strength is sufficient to drive electrons through the
collisional ions at supersonic speed in the altitude range
from 90 to 150 km. This results in excitation of the twostream instability at those altitudes. With reference to
observations in the Earth’s ionosphere, where similar
conditions typically occur in the E-region, it is suggested
that the turbulent electric fields in the two-stream
plasma waves heat the electron gas significantly. Recent
observations have shown that simultaneous to the temperature enhancement an increase of the electron density
in the E-region occurred, which can be understood as a
result of lowering of the electron–ion recombination
coefficient owing to the temperature increase. It is argued
that conditions in the Martian crustal magnetic cusp
regions allow similar processes to take place there, to
produce the electron density enhancements observed with
MARSIS.
Acknowledgments
MARSIS was built and is jointly managed by the Italian
Space Agency and NASA. Mars Express was built and is
operated by the European Space Agency. The research at
the University of Iowa was supported by NASA through
contract 1224107 with the Jet Propulsion Laboratory. We
thank all members of the ASPERA-3 team for the big
effort which led to the successful operation of the
instrument and the calibration of the data. For this paper,
we are especially grateful to Emmanuel Penou and Andrei
Fedorov at CESR for data preparation and IMA calibration and Andrew Coates at MSSL for calibration of the
ELS sensor data. We wish to acknowledge support from
German DLR grant 50QM99035, and NASA contract
NASW-00003 at SWRI. We thank the referees for
constructive comments.
ARTICLE IN PRESS
2172
E. Nielsen et al. / Planetary and Space Science 55 (2007) 2164–2172
References
Acuna, M.H., Connerney, J.E.P., Wasilevsky, P., Lin, R.P., Anderson,
K.A., Carlson, C.W., McFadden, J., 1998. Magnetic fields and plasma
observations on Mars: initial results on Mars Global Surveyor mission.
Science 279, 1676–1680.
Acuna, M.H., Connerney, J.E.P., Ness, N.F., Lin, R.P., Mitchell, D., et
al., 1999. Global distribution of crustal magnetization discovered by
the Mars global Surveyor MAG/ER experiment. Science 284, 790.
Barabash, S., Lundin, R., et al., 2006. The analyzer of space plasmas and
energetic atoms (ASPERA-3) for the Mars Express mission. Space Sci.
Rev. doi:10.1007/s11214-006-9124-8.
Bornmann, P.L., Speich, D., Hirman, J., Mathison, L., Grubb, R., Garcia,
H., Viereck, R., 1996. GOES X-ray sensor and its use in predicting
solar-terrestrial distances. Proc. SPIE 2812, 291.
Chapman, S., 1931. The production of ionization by monochromatic
radiation incident upon a rotating atmosphere, Part II. Grazing
incidence. Proc. Phys. Soc. 43, 483.
Chicarro, A., Martin, P., Traunter, R., 2004. The Mars Express mission:
an overview. In: Wilson, A. (Ed.), Mars Express: A European Mission
to The Red Planet, SP-1240. European Space Agency Publication
Division, Noordwijk, the Netherlands, pp. 3–16.
Connerney, J.E., Acuna, M., Wasilewski, P., Ness, N., Reme, H., Mazelle,
C., Vignes, D., Lin, R., Mitchell, D., Cloutier, P., 1999. Magnetic
lineations in the ancient crust of Mars. Science 284, 794.
Connerney, J.E., Acuna, M.H., Wasilewski, P.J., Kletetschka, G., Ness,
N.F., Reme, H., Lin, R.P., Mitchell, D.L., 2001. The global magnetic
field of Mars and implications for crustal evolution. Geophys. Res.
Lett. 28 (21), 4015.
Detrick, D.L., Rosenberg, T.J., Fry, C.D., 1997. Analysis of the Martian
atmosphere for riometry. Planet. Space Sci. 45 (3), 289.
Dimant, Y.S., Milikh, G.M., 2003. Model of anomalous electron heating
in the E region: 1. Basic theory. J. Geophys. Res. 108 (A9), 1350.
Duru, F., Gurnett, D.A., Averkamp, T.F., Kirchner, D.L., Huff, R.L.,
Persoon, A.M., Plaut, J.J., Picardi, G., 2006. Magnetically controlled
structures in the ionosphere of Mars. J. Geophys. Res. 111, A12204.
Fejer, B.G., Kelly, M.C., 1981. Ionosphere irregularities. Rev. Geophys.
18, 401.
Foster, J.C., Erickson, P.J., 2000. Simultaneous observations of E-region
coherent backscatter and electric field amplitude at F-region heights
with the Millstone Hill UHF radar. Geophys. Res. Lett. 27(19), 3177,
2000GL000042.
Gurnett, D.A., Kirchner, D.L., Huff, R.L., Morgan, D.D., Persoon, A.M.,
Averkamp, T.F., Duru, F., Nielsen, E., Safaeinili, A., Plaut, J.J., Picardi,
G., 2005. Radar soundings of the ionosphere of Mars. Science 310, 1929.
Gurnett, D.A., Huff, R.L., Morgan, D.D., Persoon, A.M., Averkamp,
T.F., Kirchner, D.L., Duru, F., Akalin, F., Kopf, A.J., Nielsen, E.,
Safaeinili, A., Plaut, J.J., Picardi, G., 2007. An overview of radar
soundings of the Martian ionosphere from the Mars Express spacecraft. Adv. Space Research, in press, 2007.
Lundin, R., et al., 2004. Solar wind-induced atmospheric erosion at Mars:
first results from ASPERA-3 on Mars Express. Science 305,
1933–1936.
Mendillo, M., Withers, P., Hinson, D., Risbeth, H., Reinisch, B., 2006.
Effects of solar flares on the ionosphere of Mars. Science 311,
1135.
Milikh, G.M., Goncharenko, L.P., Dimant, Y.S., Thayer, J.P., McCready,
M.A., 2006. Anomalous electron heating and its effects on the electron
density in the auroral electrojet. Geophys. Res. Lett. 33, L13809,
doi:10.1029/2006GL026530.
Morgan, D.D., Gurnett, D.A., Kirchner, D.L., Huff, R.L., Brain, D.A.,
Boynton, W.V., Acuña, M.H., Plaut, J.J., Picardi, G., 2006. Solar
control of radar wave absorption by the Martian ionosphere.
Geophys. Res. Lett. 33, L13202, doi:10.1029/2006GL026637 (J13).
Nielsen, E., 1986. Aspect angle dependence of mean Doppler velocities of
1-m auroral plasma waves. J. Geophys. Res. 91, 10173.
Nielsen, E., Schlegel, K., 1985. Coherent radar Doppler measurements
and their relationship to the ionospheric electron drift velocity.
J. Geophys. Res. 90, 3498.
Nielsen, E., Wang, X.D., Gurnett, D.A., Kirchner, D.L., Huff, R.L.,
Orosei, R., Safaeinili, A., Plaut, J.J., Picardi, G., 2007. Vertical sheets
of dense plasma in the topside Martian ionosphere. J. Geophys. Res.,
in press, 112, E02003, doi:10.1029/2006je002723.
Nielsen, E., Zou, H., Gurnett, D.A., Kirchner, D.L., Morgan, D.D., Huff,
R.L., Orosei, R., Safaeinili, A., Plaut, J.J., Picardi, G., 2006.
Observations of vertical reflections from the topside Martian ionosphere. Space Sci. Rev., doi:10.1007/s11214-006-9113-y.
Picardi, G., Biccari, D., Seu, R., et al., 2004. MARSIS: Mars advanced
radar for subsurface and ionosphere sounding. In: Wilson, A. (Ed.),
Mars Express: A European Mission to The Red Planet, SP-1240.
European Space Agency Publication Division, Noordwijk, the Netherlands, pp. 51–70.
Robinson, T.R., 1986. Towards a self-consistent non-linear theory of
radar auroral backscatter. J. Atmos. Terr. Phys. 48, 417.
Schunk, Nagy, 2000. Ionospheres. Cambridge University Press, New York.
St.-Maurice, 1990. Electron heating by plasma waves in the high latitude
E-region and related effects: observations. Adv. Space Res. 10 (6),
239.
St.-Maurice, J.P., Schlegel, K., Banks, P.M., 1981. Anomalous heating of
the polar E region by unstable plasma waves. II–Theory. J. Geophys.
Res. 86, 1453.
St.-Maurice, Kofman, W., kluzek, E., 1990. Electron heating by plasma
waves in the high latitude E-region and related effects: observations.
Adv. Space Res. 10 (6), 225.
Sudan, R.N., Akinrimisi, J., Farley, D.T., 1973. Generation of small scale
irregularities in the equatorial electrojet. J. Geophys. Res. 78, 240.
Withers, P., Mendillo, M., Risbeth, H., Hinson, D.P., Arkani-Hamed, J.,
2005.
Ionospheric
characteristics
above
Martian
crustal
magnetic anomalies. Geophys. Res. Lett. 32, L16204, doi:10.1029/
2005GL023483.